O-241 Association of a deep learning-based scoring system with morphokinetics and morphological alterations in human embryos

Abstract

Abstract Study question How is the deep learning-based scoring system, iDAScore, associated with biological events during the pre-implantation period? Summary answer The morphokinetic analysis revealed that during the transformation to blastocyst stage, morphokinetic and morphological events were strongly associated with iDAScore. What is known already A recent study reported that the commercially available deep learning model for automatic scoring, iDAScore, performed as well as, or even better than, the more traditional embryo assessment-based model on blastocyst morphology or annotation-dependent ranking tools. Furthermore, a correlation between iDAScore and the incidence of rapid cleavage, time duration until the blastocyst stage, and morphological grade of the inner cell mass and trophectoderm was reported. However, the association of iDAScore with other key morphokinetic parameters remains undetermined. Study design, size, duration This retrospective observational study included 925 patients who underwent oocyte retrieval in a clomiphene citrate-based minimal stimulation cycle and obtained expanded blastocysts in ICSI cycles between October 2019 and December 2020. Oocytes of patients with different diagnoses of infertility were included in the analysis, while cases involving cryopreserved gametes or surgically retrieved sperm were excluded. The association of iDAScore with morphokinetics and morphological alteration during fertilisation, cleavage stage, compaction, and blastocyst stage were analysed. Participants/materials, setting, methods Microinjected oocytes were assessed using a time-lapse monitoring (TLM) culture system (Embryoscope). Oocytes not suitable for TLM assessment, due to excess of residual corona cells or inadequate orientation for correct observation, were not analysed. Phenomena, relevant to meiotic resumption, pronuclear dynamics, cytoplasmic/cortical modifications, cleavage pattern, and embryo quality, were annotated. Multiple linear regression analysis was used to assess the relative importance of the possible predictor variables in explaining the iDAScore. Main results and the role of chance The duration of the cytoplasmic halo was significantly prolonged in the low scoring blastocysts. The timing of female and male pronuclei breakdown was significantly delayed in the low scoring blastocysts compared with the high scoring blastocysts. Embryos with either trichotomous, multi-chotomous, rapid, or reverse cleavage or asymmetric division exhibited a lower score than embryos with normal cleavage. The cell number and amount of blastomere fragmentation on days 2 and 3 were significantly associated with iDAScore. Delayed division, compaction, blastulation, and blastocyst expansion were observed in the low scoring embryos. The incidence of blastomere exclusion and extrusion during embryonic compaction was significantly higher in low scoring embryos than in high scoring embryos. Blastocyst morphology was significantly associated with iDAScore. Multiple linear regression analysis revealed that morphokinetic and morphological events were strongly associated with iDAScore especially during the transformation to blastocyst stage, which has been considered important in embryo assessment. It was also revealed that some morphological parameters and time durations during the cleavage stage were also correlated with the iDAScore. Limitations, reasons for caution The study data derive from treatments carried out in a single centre. The study findings therefore require independent verification from other research groups. Wider implications of the findings We found that iDAScore was significantly correlated with morphokinetics and morphological alterations of pre-implantation embryos, especially during the late pre-implantation period. Our findings may contribute to the literature on deep learning model-based embryo selection, which may provide patients with a compelling explanation regarding blastocyst selection. Trial registration number not applicable